Journal of Food Engineering 89 (2008) 488–497

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Optimum design and operating conditions of multiple effect evaporators:

Tomato paste
R. Simpson *, S. Almonacid, D. López, A. Abakarov
Departamento de Procesos Químicos, Biotecnológicos, y Ambientales, Universidad Técnica Federico Santa María, Casilla P.O. Box 110-V, Valparaíso, Chile

a r t i c l e i n f o a b s t r a c t

Article history: Agro industry is a very important industrial sector worldwide, especially for countries like New Zealand
Received 22 January 2008 and Chile. The main objective of this research was to propose a new economic evaluation procedure to
Received in revised form 24 May 2008 optimize the design and operation of multiple effect evaporators and compare it with the traditional
Accepted 27 May 2008
chemical engineering approach of total cost minimization. The proposed strategy incorporates a quality
Available online 10 June 2008
factor expressed as a function of lycopene concentration on the final product to find the optimal number
of effects and operating conditions through the maximization of the net present value.
Keywords:
The mathematical model was implemented using Microsoft Excel and considered mass and energy bal-
Multi-effect evaporators
Economic evaluation
ances, specific relations for tomato concentration and a first order degradation kinetic for lycopene. The
Net Present Value results indicate that when augmenting the capacity of the evaporation system of 5 effects from 50 to
Quality 75 Ton/h, the lycopene retention increases from 95.25% to 96.27%. When evaluating the system through
Lycopene the logic of the total cost minimization, an optimum of 4 effects is found, but when evaluating the system
Process optimization using the maximization of the Net Present Value including lycopene as a quality parameter, the optimum
is 3 effects.
It appears of extreme relevance to consider quality as an intrinsic and integral part of the process
design, as it will then be possible to identify several potential improvements in different food processes.
Ó 2008 Published by Elsevier Ltd.

1. Introduction greater rewards. Questions as to just what should be maximized

Process optimization has always been a noble objective of engi-
neers entrusted with the responsibility for process development
and improvement throughout the food industry. Examples of
sophisticated mathematical approaches to process optimization,
in which some objective function is maximized or minimized
subject to chosen constraints, are widely published in literature
(Douglas, 1988). On the other hand, the chemical industry has used
cost analysis in several cases in relation to design and process opti-
mization. A classical example in the chemical industry is the deter-
mination of the optimal number of effects in a evaporation system,
were the optimum is found when there is an economic balance be-
tween energy saving and added investment, this is, a minimization
of the total cost (Kern, 1999). In this vision, although correctly, qual-
ity is not considered as a parameter in the determination of the opti-
mum number of effects, so the process specifications and operating
conditions are assumed independent of both product quality as its
sale price. The purpose of this manuscript is to suggest that the
extrapolation of optimization problems from chemical industry to
the food industry may often be restricted to an unnecessarily nar-
row or local domain, and that a more global perspective may reap
* Corresponding author. Tel.: +56 32 2654302; fax: +56 32 2654478.
E-mail address: ricardo.simpson@usm.cl (R. Simpson).
or minimized, or what are the real constraints, as opposed to only
those that are immediately apparent, are questions often posed
without a broad enough view of the ‘‘big picture”.
China continues to make inroads in the world market. China is
the world’s largest tomato products producer and exporter, fol-
lowed by the EU and the United States. China produces tomato
products mainly for the export market, with exports accounting
for more than 85% of production. Since calendar year (CY) 1999,
China’s tomato paste exports have had an average annual increase
of 30% (USDA, 2007).
The production of tomato paste is highly seasonal, and then,
maximizing production levels in this industry is of vital impor-
tance. The process is generally done in multiple evaporation sys-
tems, with a different number of effects, through which the
content of water is diminished until a final concentration from
30 to 32 °Brix is acquired, and where temperatures generally do
not exceed 70 °C.
Lycopene is the main carotenoid found in tomatoes and many
studies have showed its inhibiting effect on carcinogenic cell
growth (Shi et al, 2007). It is also the component which generates
the red characteristic color in tomatoes, among other fruits and
vegetables (Goula and Adamopoulos, 2006). A study developed
by the University of Harvard, revealed that the consumption of
lycopene reduced the probabilities of generating prostate cancer

0260-8774/$ - see front matter Ó 2008 Published by Elsevier Ltd.

doi:10.1016/j.jfoodeng.2008.05.033
 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497
Nomenclature
A heat transfer area, m2
Cu cost per unit, US$/kg
Cp specific heat of concentrate (kJ kg1 °C1)
Deb boiling point elevation (BPE) or Boiling point rise (BPR)
(°C)
E activation energy (kJ/mol)
F mass flowrate (kg/h)
H enthalpy (kJ/kg)
i annual interest rate
I total investment (US$)
k reaction rate constant (1/h)
k0 frequency factor (1/h)
K0 constant
M mass in the evaporator (kg)
m project shelf life, years
n number of effects
NPV net present value (US$)
O evaporation system operation
by 45%, in a population of 48,000 subjects who had at least 10 ra-
tions of tomatoes or sub products in their weekly diet. Other re-
search discovered that lycopene also reduces cholesterol levels in
the form of a lipoprotein of low density (LDL), which produces ath-
erosclerosis; this means that the consumption of tomatoes reduces
the effects produced by cardiovascular diseases.
Lycopene as the main organic compound presents a denatural-
ization reaction rate that is time and temperature dependent. Then,
for the mathematical model of the behavior of multi-effect evapo-
rators, it is very important to have a good overview of the general
fluctuation of lycopene retention or loss under different system de-
signs and operating conditions.
As aforementioned, most food processes have been adapted and
extrapolated from the chemical engineering industry without an
adequate consideration of product quality during system design
and process optimization. That is certainly a good start, but maybe
somewhat limited and might have inhibited us to take a more glo-
bal view. For example, it appears of extreme relevance to consider
quality more frequently as an intrinsic and integral part of process
design. In the food industry, the main effort is commonly related to
the maximization of the quality of the product, which is not neces-
sarily the case in the chemical industry. Generally the optimization
of food processes have been restricted to determining the optimal
operating conditions of an allegedly, well designed food process.
Nevertheless, if quality is considered as a parameter in the system
design, it is very probable that the new design will differ from the
original one.
For example, in the case of a multiple effect evaporator system
for the processing of tomatoes, the optimization of the design is
only focused on an economic analysis which combines the invest-
ment (number of effects) and the operating costs (steam consump-
tion) (Kern, 1999). This strategy does not include quality as an
integral part of the economic evaluation, even though previous
studies have demonstrated the dependence of the final product
price towards quality of the final product (Schoorl and Holt, 1983).
The main objective of this research work is to propose a new
economic evaluation procedure to optimize the system design
and operation of tomato juice, multiple effect evaporator and com-
pare it to the traditional chemical engineering approach of total
cost minimization. The proposed strategy will incorporate a quality
factor which will be expressed as a function of lycopene concentra-
tion on the final product to find the optimal number of effects and
operating conditions through the maximization of the Net Present
Value (NPV).
489
P pressure (kPa)
Pu unit sales price (US$/kg)
Q heat flow (kJ)
Q*j annual production at period j (units/year)
R ideal gases constant (CkJ/mol)
T temperature (°C)
t time (h)
X concentration of soluble solids, kg ss/kg
Y lycopene concentration kgL/kg ss
Subscripts
c condensing
d download
e cooling
i evaporator effect, i
j evaluation period, j
p losses
2. Methodology
2.1. Problem description
Cost analysis has been extensively and correctly utilized in find-
ing the best process design in several chemical engineering plants.
A classical example is multi-effect distillation. In this case, cost
analysis should aim to determine the optimum number of effects
in multiple-stage equipment. According to the literature in chemi-
cal engineering ‘‘The optimum number of effects must be found from
an economic balance between the savings in steam obtained by multi-
ple effect operation and added investment”. It is important to eluci-
date whether the aforementioned approach is recommendable in
the optimization of food processes? From a microeconomics point
of view, this approach is correct but, it is important to consider that
the different equipment configurations are producing exactly the
same quality of the end product. Meaning that, independent of
the number of effects, not only will we be able to reach the same de-
gree of concentration, but also the same quality. In addition, by
changing the equipment configuration it is possible to attain the
same degree of concentration but with different product quality.
At least, for multi-effect evaporation in food processing, the re-
ferred approach is not necessarily the right micro-economic tool
to find the optimum number of effects. For this kind of application,
a correct micro-economic analysis should consider not only all
costs but also the expected benefits. According to the relevant tech-
nical literature, an adequate micro-economic procedure is to max-
imize the NPV.
In the case of the evaporation process for tomato paste, the pro-
duction is highly seasonal, and, in addition, the product quality
could be highly affected by the operating conditions. Therefore, it
is important to consider the impact of the installed capacity and
the final product quality. In the following, we will compare two dif-
ferent economic approaches: (a) determination of the optimal
number of effects by minimizing the total cost and, (b) maximiza-
tion of the NPV, considering quality as an intrinsic parameter of the
modeled system.
2.2. Product quality
To reach the objective of the present research work, a quality
parameter must be considered to the mathematical model of the
evaporation system. The chosen parameter is lycopene, because,
490 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497

as mentioned before, this carotenoid pigment is what gives toma- system that must be solved (see Fig. 1 for a schematic representa-
toes their characteristic color, and, in addition, it has some medical tion of the system), operates on countercurrent and the total num-
benefits. ber of effects varies from 1 to ‘‘n”. The value of ‘‘n” and the
Usually, degradation rates in sensitive food components are operating conditions will be determined at the end of this work
modeled as a first order kinetic, as follows: through the maximization of the NPV.
The total mass balance in evaporator effect ‘‘i” is:
dY
r ¼ kY or ¼ kY ð1Þ
dt dðMi Þ
¼ F iþ1  Fvi  F i ð5Þ
The Arrhenius equation relates specific reaction rate constant to dt
temperature according to: If the mass within the evaporator effect is controlled, then, un-
  der steady state, Eq. (5) can be written as:
E
k ¼ k0  exp  ð2Þ
RT 0 ¼ F iþ1  Fvi  F i ð6Þ

The first order kinetic for lycopene degradation has been con- In the same way, a mass balance for soluble solids at effect i, can
firmed by Goula and Adamopoulos (2006). In the same research be written as:
study an equation was obtained to determine the reaction rate in dðMi X i Þ
the lycopene degradation, as a function of temperature and soluble ¼ F iþ1  X iþ1  F i  X i ð7Þ
dt
solids concentration X expressed in °Brix.
Under steady state condition:
For X  55; K ¼ 0:121238 expð0:0188XÞ
  0 ¼ F iþ1  X iþ1  F i  X i ð8Þ
2317 1
 exp  ðmin Þ ð3Þ
T þ 273:15 The corresponding energy balance for the evaporator effect i, is:
dðHi Mi Þ
¼ F iþ1  Hiþ1 þ Fvi1  Hvi1  F i  Hi  Fvi  Hvi  Fci  Hci þ Q p
dt
ForX  55; k ¼ 0:275271 exp ð0:00241X Þ
  ð9Þ
2207 1
 exp  ðmin Þ ð4Þ Under steady state condition:
T þ 273:15
0 ¼ F iþ1  Hiþ1 þ Fvi1  Hvi1  F i  Hi  Fvi  Hvi  Fci  Hci þ Q p
In our research study, the system to be modeled should con-
sider tomato concentration in the range of 5–35 °Brix, so only Eq. ð10Þ
(4) will be required.
The enthalpy of the tomato paste was estimated through the
specific heat (Cp), utilizing the following expression (Tonelli et
2.3. Model development
al., 1990):

The evaporation process involves mass and heat transfer (Him-
melblau and Bischoff, 1968). The tomato juice was considered as a
binary solution of water and soluble solids, both considered inert
in a chemical sense. Under these considerations, one effect of the
industrial evaporator can be shown in the manuscript by Miranda
and Simpson, 2005.
So the macroscopic model is of the knowledge-type based on
conservation laws and also empirical relationships which describe
the equilibrium phases. These relationships have been rearranged
from non-linear algebraic equations from literature, with the expe-
rience taken from the experimental site. Only the juice phase is
considered for modeling.
The modeling assumptions are:
– Homogenous composition and temperature inside each
evaporator.
– Constant juice level in each evaporator.
– Thermodynamic equilibrium (liquid–vapor) for the whole mod-
eled system.
The mathematical model developed in this research study in-
cluded specific relationships for lycopene degradation. The general
Hi ¼ ð4:184  2:9337  X i Þ  T i ð11Þ
The following thermodynamic relationship describes the boil-
ing point rise (BPR) or boiling point elevation (BPE), whose param-
eters have been determined experimentally, it is one of the three
important properties (specific heat, viscosity and boiling point
rise), that must be specified in a multiple effects evaporator (Rizvi
and Mittal, 1992). This property (BPR) is significant at high soluble
solids concentration. On a multiple effect equipment, the effective
temperature differences decrease for the combination of boiling
point. The following correlation reported by Miranda and Simpson,
2005, was utilized.
Deb ¼ 0:175X 1:11 e3:86X P0:43 ð12Þ
Vapor was considered saturated within the evaporator. The fol-
lowing correlations were obtained from Perry and Chilton (1973),
and allow for the estimation of vapor properties with an error of
less than 1%.
For 40 °C < Tv < 70 °C.
Tvi ¼ 32:5515  P0:2898
i  17:7778 ð13Þ

Fig. 1. Evaporation system of n effects operated in countercurrent.

 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497 491

For 70 °C < Tv < 135 °C. From Eq. (21) two significant terms can be distinguished total
investment (I); and annual benefits (bj). Where the investment
Tvi ¼ 39:0514  P0:2382  17:7778 ð14Þ
i – for a given capacity – can be expressed as a function of the num-
For 40 °C < Tv < 135 °C. ber of effects (NE), and benefits are related to the product’s unit-
price (Pu) and costs per unit (Cu).
Hvi ¼ 2509:2888 þ 1:6747  Tvi ð15Þ
bj ¼ Q j ðP u  C u Þ ð22Þ
Hci ¼ 4:1868  Tci ð16Þ
Clearly, in food products, unit-price is not constant and it is di-
To estimate lycopene degradation (or retention) in each evapo- rectly related to the final quality. As aforementioned, product qual-
rator effects, a mass balance at effect i was carried out as follows: ity is related to the number of effects (NE), in general with process
  arrangement (PA) and operating conditions (OC). In addition, the
dðX i Y i Þ dðM i X i Y i Þ
F iþ1  X iþ1  Y iþ1  F i  X i  Y i þ M i ¼ ð17Þ incidence of energy in unit-cost can be expressed as a function of
dt dt
the number of effects (NE) and operating conditions (OC) too.
Assuming steady state, perfect mixing and first order lycopene Therefore Eq. (21) can be expressed as a function of the numbers
degradation rate, the following expression is obtained. of effects and operating conditions, as follows:
F iþ1  X iþ1  Y iþ1  F i  X i  Y i  M i  X i  ki  Y i ¼ 0 ð18Þ X
n
Q j ðPu ðNE ; OCÞ  C u ðN E ; OCÞÞ
NPV ¼ IðN E Þ þ ð23Þ
Solving for Yi. j¼1 ð1 þ iÞj
F iþ1  X iþ1  Y iþ1 For fixed operating conditions, to find the critical value for the
Yi ¼ ð19Þ
F i  X i þ M i X i ki number of effects (NE for maximum NPV), it is necessary to derive
Eq. (23) and then equalize it to zero, so N E can be obtained (critical
Combining Eqs. (4) and (19) the following equation is obtained.
value). On the other hand, critical value N E will represent a maxi-
F iþ1  X iþ1  Y iþ1 mum for NPV if the second derivative is smaller than zero.
Yi ¼ 2207
F i  X i þ Mi  X i  ð0:275271  expð0:00241  X i Þ  expð T þ273:15 ÞÞ !
i
dðNPVÞ dðIðNðEÞÞ d Xn
Q j ðPu ðN E Þ  C u ðN E ÞÞ
ð20Þ ¼ þ ð24Þ
dN E dNE dNE j¼1 ð1 þ iÞj
With the previous equation (Eq. (20)), lycopene concentration for
the output flowrate in each evaporator effect can be estimated Considering:
knowing the steady state values, the mass inside each evaporator b1 ¼ b2 ¼    ¼ bn ¼ b ¼ Q  ðPu ðNE Þ  C u ðNE ÞÞ ð25Þ
effect and the lycopene concentration in the input flowrate of each
evaporator effect. It is important to mention that when trying to Expressing the annual benefits as a present value, the second
estimate lycopene concentration, a degree of freedom is added to term of the right hand side of Eq. (21) can be reduced to:
the system, which is satisfied with the data of lycopene concentra- X
n
bj X
n
Q  ðPu ðNE Þ  CðNE ÞÞ
tion in the evaporator system input (feeding) flowrate. j
¼ ¼ K 0 ðPu ðNE Þ  C u ðNE ÞÞ ð26Þ
j¼1 ð1 þ iÞ j¼1 ð1 þ iÞj

2.4. Economic evaluation Therefore, replacing it in Eq. (24), we obtain:

dðNPVÞ dðIðN E ÞÞ dðPu ðN E Þ  C u ðNE ÞÞ
Firstly, we shall do a simple and preliminary analysis to find an ¼ þ K0 ð27Þ
dN E dNE dNE
optimum number of effects when processing a food product. The
idea is to do a direct comparison between total cost minimization
dðNPVÞ dðIðN E ÞÞ dðPu ðN E ÞÞ dðC u ðNE ÞÞ
and NPV maximization. As stated earlier, the focus of the economic ¼ þ K0  K0 ð28Þ
dN E dNE dNE dNE
analysis will be the inclusion of quality as an intrinsic parameter of
the process design. In the case of a typical chemical engineering analyses, quality is
Generally, in the chemical industry, the quality of the product considered to be independent of the number of effects (NE). Mean-
to be concentrated on is associated with the final concentration ing that the unit-price (Pu) is constant (independent of NE), there-
of the product independent of equipment design and operating fore Eq. (28) can be expressed as:
conditions. It is for this reason, that the economic evaluation
dðNPVÞ dðIðN E ÞÞ dðC u ðNE ÞÞ
and optimization of these processes have been based on a total ¼  K0 ð29Þ
dN E dNE dNE
cost minimization. However, clearly, for food processes, product
quality is highly dependent on equipment design and operating By inspection of Eq. (29) it is clear that the maximum NPV value is
conditions. the same as the minimum total cost. This is the reason why, in most
In the particular case of tomato paste, the most important com- chemical engineering analyses the search for the optimum process
ponent together with the compliance of concentration and consis- is reduced to find the minimum total cost. Quantification of NPV
tency is the final content of lycopene in the product. and total cost was mainly due by literature references and some di-
rect quotations (Maroulis and Maroulis, 2005; Peters et al., 2003).
2.5. Maximization of NPV and minimization of total cost In food processing, quality has a strong effect on product price
and, in addition, has a long term effect on the consumer’s percep-
A criterion to optimize process design is to determine the num- tion of the food company. In this case, but also in general, the opti-
ber of effects that shall maximize the NPV of the invested capital mum process for a specific technology can be obtained by
for the new process line. This can be approached on the basis of implementing and managing Eq. (22).
microeconomics.
Where, the following equation is the expression for NPV: 3. Results and discussions
X
n
bj
NPV ¼ I þ ð21Þ The mathematical model was developed for evaporator systems
j¼1 ð1 þ iÞj from 1 to 7 effects, operated under countercurrent. In each system,
492 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497

the restriction was equal areas for each evaporator effect. The input the final product is affected as a function of the residence time in
values to the model, shown in Table 1, were the same for all of the the evaporation system. While the number of effects increases in
systems and obtained from an actual industrial plant, comple- the evaporation system, the total residence time increases as well
mented with available online information from manufacturers. as the temperature at which the product is exposed, and, therefore,
there is also an increase of the lycopene degradation.
3.1. Steady state conditions Lycopene retention for the theoretical evaporation system for
tomato paste are shown in Fig. 4. Clearly, lycopene retention de-
From the mass and energy balance equations, liquid–vapor creases as the number of effects increase, which is justified by
equilibrium equation and specific relations for the tomato paste, the augmentation of the total residence time of the system and also
a steady state model for the evaporator system was developed, because of the temperature rise at which the tomato paste is ex-
considering one up to seven effects. From this information, it is posed to. In regards to the supply flowrate, there is an increase
possible to verify the decrease in vapor flowrate necessary for in the lycopene retention when augmenting the flowrate. The in-
the operational process and an increase in the total system area, crease in lycopene retention is less abrupt when the supply flow-
when augmenting the number of effects (Fig. 2). rate is over 100 ton/h.
To have a more precise view of the product behavior in the
evaporation system, residence time and their respective tempera- 3.3. Changes in processing capacity
tures, are presented in Table 2 for each one of the effects in the dif-
ferent systems. To have a more precise idea of the effect of temperature and res-
idence time effect on lycopene degradation process capacity was
3.2. Lycopene retention set to different values in a specific range. This was done by main-
taining the heat exchange area, and consequently, the number of
Lycopene retention in the final product was estimated for each effects (a 5 effect lineup system was considered, as it is the number
one of the alternative systems from the data obtained under steady that is regularly used in the tomato paste industry). Increasing the
state operation. From the results shown in Fig. 3, it is clear that processing capacity of the system, results in a increment of the
lycopene concentration in the final product has a linear decay required energy, therefore steam inlet pressure Pv0 will also be
when augmenting the number of effects in the evaporation system. increased. This also implies a rise of temperature in the evapora-
The previous result gives a clue of how the content of lycopene in tor’s effects. As shown in Fig. 5, the required vapor flowrate
increases proportionally to the evaporation system input flowrate,
steam inlet pressure increases in a second order polynomial way.
Table 1
Fig. 6 shows a decrease in the systems total residence time as well
Input data for mathematical model implementation
as an increase in each evaporator’s temperature.
Name Variable Value As it was expected, temperature inside each evaporator in-
Input flowrate FAl, kg/h 50,000 creases due to the augmentation of the system energy require-
Input temperature TAl, °C 98 ments. The systems residence time decreases because of the
Initial soluble solids XAl, kg ss/kg 0.05 increase of input flowrate and the conservation of the holdup va-
concentration
Input concentration YAl, kg Lic/kg SS 0.01
lue. As it is observed in Fig. 7, there is an increase in lycopene
of lycopene retention when augmenting the input flowrate associated to the
Final soluble solids X1, kg ss/kg 0.3 products residence time which decreases in the evaporation sys-
concentration tem, with no regard to the increase of the evaporator temperature
Steam inlet pressure Pv0, kPa 143.4
increment.
Temperature change Tvn  Td, °C 2
in condensator This is of great importance, because it demonstrates that lyco-
Operation pressure in Pn, kPa 16.5 pene is not an obstacle to increase the processing capacity of the
evaporator ‘‘n” evaporation system, therefore the maximum capacity will only

Fig. 2. Total transfer area m2 and steam inlet flowrate ton/h vs. number of effects.
 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497 493

Table 2
Temperature data °C and residence time h for systems from 1 to 7 effects with an input flowrate of 50 ton/ha,b
Effect Number of effects in the system
number
1 2 3 4 5 6 7
Residence Temp. Residence Temp. Residence Temp. Residence Temp. Residence Temp. Residence Temp. Residence Temp.
time time time time time time time
1 1.10 55.9 0.23 55.6 0.17 55.6 0.15 55.6 0.14 55.6 0.13 55.6 0.13 55.6
2 0.84 76.8 0.28 67.5 0.20 63.8 0.17 61.8 0.15 60.5 0.14 59.6
3 0.78 84.7 0.31 74.3 0.22 69.3 0.19 66.3 0.17 64.3
4 0.75 89.3 0.34 78.9 0.24 73.3 0.20 69.8
5 0.73 92.3 0.36 82.3 0.26 76.4
6 0.71 94.4 0.38 84.8
7 0.70 96.1
1 effect 2 effects 3 effects 4 effects 5 effects 6 effects 7 effects
Output pressure at each effect
P1, kPa 16.5 40.18 56.4 67.7 76.05 82.41 87.5
P2, kPa 16.5 27.7 36.5 44.76 51.6 57.31
P3, kPa 16.5 23.8 29.75 34.91 40.15
P4, kPa 16.5 21.82 26.38 30.32
P5, kPa 16.5 20.65 24.28
P6, kPa 16.5 19.86
P7, kPa 16.5
a
Feed enters effect 1.
b
Fresh vapor pressure: 143.4 kPa.

Fig. 3. Lycopene retention% vs. number of effects for an input of 50 ton/h.

Fig. 4. Lycopene retention% for a system from 1 to 7 effects.

494 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497

Fig. 5. Changes in energy consumption represented by the vapor flowrate kg/h and steam inlet pressure kPa as a function of the input flowrate ton/h.

Fig. 6. Residence time (min) and temperature °C as a function of the input flowrate ton/h.

Fig. 7. Lycopene retention% as a function of the input flowrate ton/h.

 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497 495

Fig. 8. Cost evaluation for an evaporator system with an input flowrate of 50 ton/h.

be restricted by available vapor pressure, minimum specific hold-
up, pumps power among others.
3.4. Economic evaluation
The economic evaluation consists of determining the optimum
number of effects and operating conditions of the system. The eco-
nomic evaluation of the system was done in two different ways.
Firstly, an economic evaluation with the concept of minimizing
the total costs, and secondly, an economic evaluation to maximize
the NPV taking into account the impact of the process design and
operating conditions on product quality.
3.5. Optimum number of effects
The economic evaluation was carried out by simple inspection.
This is where the steady state conditions for systems with 1–7 ef-
fects were found, and then total cost minimization and NPV max-
imization methodologies were used. The search was focused to
find the number of effects that minimize the total cost and, in addi-
tion, to find the number of effects that maximize the NPV.
The results for each evaluation systems are shown in Figs. 8
and 9. The total cost minimization (Fig. 8) shows an optimum of
4 effects. Nevertheless, when doing NPV maximization (Fig. 9),
the number of optimum effects was 3 due to the inclusion of the
quality parameter on the evaluation procedure. Naturally, for dif-
ferent processing capacities, the optimum number of effects varies
for both evaluation procedures. This is why differences are encoun-
tered in the optimum number of effects in some operation ranges.
In Fig. 10, the optimum number of effects is presented for different
operation ranges. As it is observed in Fig. 10, when evaluating the
evaporation system, including the quality parameter, in the range
of 25 ton/h through 50 ton/h, the optimum number of effects de-
creases, in comparison to the evaluation done based on total costs
only. This is explained with previous results where a decrease of
lycopene retention was a result of the increase of the number of ef-
fects. It is for this reason that the NPV maximization, in this partic-
ular case tends to be a lower number of effects.
3.6. Optimum operating conditions
In the search for the optimum operating conditions of evapora-
tion system, the system was economically evaluated under a vari-
able steam inlet pressure (Pv0) where the inclusion of lycopene as a
quality parameter was considered. As a constraint to the problem,
it was estimated that the output temperature of the tomato paste

Fig. 9. Net Present Value evaluation for evaporator systems with input flow of 50 ton/h.
496 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497

Fig. 10. Optimum number of effects for different input flow rates ton/h according to total cost minimization and Net Present Value maximization.

Table 3 (highest temperature), with the optimum number of effects for a

Optimum operating conditions defined vapor pressure, could not be higher than 95 °C. Results ob-
1st effect 2nd effect 3rd effect tained shown an increase in the project profitability when steam
inlet pressure is augmented. Optimum operating conditions, with
Heat transfer area, m2 209.6 209.2 209.1
Global heat transfer coefficient, 4494.5 6767.2 8475.4
the previous stated constraints, is found to be a 3 effects system
kJ/°C/m2/h with a steam inlet pressure of 260 kPa (see Table 3). This result dif-
Heat transferred, MJ/h 32803.8 30941.1 28727.7 fers from that obtained from the evaluation based on the total cost
Boiling point raise, °C 0.53 0.07 0.03 minimization (5 effects system).
Holdup, kg 5320 4907 4815
Table 4 depicts the results obtained by both economics ways;
Residence time, h 0.639 0.224 0.141
DT Nukiyama 34.8 21.9 16.2 total cost minimization and NPV maximization.
Steady state values
Steam inlet flowrate, kg/h 15012.9 4. Conclusions
Steam inlet pressure, kPa 260
Steam inlet temperature, °C 129.1 The steady state values of the evaporator system were able to
Steam inlet enthalpy, kJ/kg 2725.45
be linked to the reaction kinetics of the target attribute, lycopene.
Output flowrate, kg/h 8322.5 21930.0 34265.4 A mathematical model was successfully developed, and then an
Temperature, °C 94.26 71.88 55.60
Concentration SS, kg/kg 0.300 0.114 0.073
economic evaluation of the optimum design and operating condi-
Lycopene concentration, kg/kg SS 0.0096790 0.0099130 0.0099724 tions of the evaporation system (1–7 effects operated under coun-
% Lycopene retention 96.79% 99.13% 99.72% tercurrent) was carried out.
Concentrate enthalpy, kJ/kg 311.3 276.7 220.7 It was possible to determine that the lycopene retention has a
Vapor flowrate, kg/h 13607.5 12335.4 15734.6 linear decay respect to the number of effects used in the evapora-
Vapor pressure, kPa 81.84 32.9 16.5 tion system.
Vapor temperature, °C 93.73 71.81 55.57
When analyzing the behavior of a 5 effect evaporator system, an
Vapor enthalpy, kJ/kg 2666.3 2629.6 2602.4
increase in the processing capacity from 50 ton/h to 75 ton/h aug-
Condensed flowrate, kg/h 15012.9 13607.46 12335.4
ments the lycopene retention in the final product from 95.25% to
Condensation temperature, °C 129.08 93.7 71.81
Condensed enthalpy, kJ/kg 540.4 392.4 300.7 96.27%.
The previous result is due to the decrease in residence time, and
independent of an increase in the evaporator’s temperature. This
result is important because it indicates that the increment in toma-
to paste production, in this particular case, is restricted only by
mechanical factors like available vapor pressure, maximum speci-
Table 4
Optimum number of effects for different steam inlet pressures fied holdup, and pump power, etc.
The total cost minimization allows the determination of the
Pv0, kPa Number of optimum effects
‘‘best” equipment design (optimum number of effects), but no
Minimum total cost Maximum NPV information on operating conditions and product quality is ob-
110 4 2 tained. On the other hand, with the NPV approach, it is possible
120 4 2 to optimize the system design and operating conditions simulta-
130 4 3 neously. In addition, the NPV approach considers the final product
140 4 3
150 4 3
quality as an intrinsic parameter of the system.
160 4 3 With the inclusion of lycopene as a quality parameter (NPV), the
170 4 3 optimum number of effects decreases from 4 to 3 when compared
200 4 3 with total cost analysis. In addition, it was also possible to deter-
230 4 3
mine the optimum operating conditions of the 3 effects system
260 5 3
at 260 kPa.
 R. Simpson et al. / Journal of Food Engineering 89 (2008) 488–497
It appears of extreme relevance to consider quality as an intrinsic
and integral part of the process design, as it will then be possible to
identify several potential improvements in different food processes.
Acknowledgements
Author Ricardo Simpson is grateful for the financial support
provided by CONICYT through the FONDECYT project number
1070946. Author Sergio Almonacid is grateful for the financial sup-
port provided by CONICYT through the FONDECYT project number
1070512.
References
Douglas, J.M., 1988. Conceptual Design of Chemical Processes. Chemical
Engineering Series. McGraw-Hill International Editions.
Goula, A., Adamopoulos, K., 2006. Prediction of Lycopene degradation during a
drying process of tomato pulp. J. Food Eng. 74 (1), 37–46.
497
Himmelblau, D.M., Bischoff, K.B., 1968. Process analysis and simulation:
deterministic system. Wiley, New York, USA.
Kern, D. 1999. Procesos de Transferencia de Calor. Editorial Continental S.A. Mexico.
Maroulis, A.Z., Maroulis, Z.B., 2005. Cost data analysis for the food industry. J. Food
Eng. 67 (3), 289–299.
Miranda, V., Simpson, R., 2005. Modeling and simulation of an industrial multiple
effect evaporator: tomato concentrate. J. Food Eng. 66 (2), 203–210.
Perry, J., Chilton, C., 1973. Chemical Engineers Handbook. Editorial Mc Graw-Hill,
USA.
Peters, M., Timmerhaus, K., West, R., 2003. Plant Design and Economics for Chemical
Engineers. Editorial Mc Graw-Hill, USA.
Rizvi S.H., Mittal G.S. 1992. In: Van Nostrand Reinhold (Ed.), Experimental Methods
in Food Engineering..
Schoorl, D., Holt, J.E., 1983. An analysis of the effect of quality on prices of
horticultural procedure. Agricultural Systems 12 (2), 75–99.
Shi, J., Qu, Q., Kakuda, Y., Jun, S., Jiang, Y., Koide, S., Shim, Y., 2007. Investigation of
the antioxidant and synergistic activity of Lycopene and other natural
antioxidants using LAME and AMVN model systems. J. Food Compos. Anal. 20
(7), 603–608.
Tonelli, M., Romagnoli, J., Porras, J., 1990. Computer package for transient analysis of
industrial multiple-effect evaporators. J. Food Eng. 12 (4), 267–281.
USDA, 2007. Tomatoes and Tomato Products. World Markets and Trade.